Hypertension (HTN) is the leading preventable risk factor for cardiovascular disease (CVD) and premature death worldwide (1, 2). The prevalence of HTN is rising globally, especially in low- and middle-income countries (3). In 2010, an estimated 1.39 billion people, equaling 31.1% of the global adult population, had HTN, defined as systolic blood pressure (BP) ≥ 140 mm Hg, diastolic BP ≥ 90 mm Hg, and/or current use of antihypertensive medication (3). The prevalence of HTN was higher in men than in women and was lower in high-income countries than in low and middle-income countries (3). Rural residents had a higher prevalence of HTN than urban residents in high and middle-income countries, but the trend was reversed in low-income countries (3). These differences in prevalence are not fully understood but are likely associated with differences in disease awareness, unhealthy diet, physical activity and healthcare resources at different urbanization levels and socioeconomic statuses (3). Heart failure (HF) is also an increasing global pandemic, affecting at least 64.3 million people worldwide (4). HTN is important risk factor for HF and is comorbid in most patients with HF (4, 5). A previous study showed that 20 mmHg elevation of systolic BP is associated with 50% increased risk of HF (6). In the Framingham Heart Study cohort, 91% of the participants developed HTN before HF, and compared with normotensive individuals, the male and female patients with HTN had 2- and 3-fold increased risks of HF, respectively (7). Elevated BP changes the structure and function of blood vessels and the left ventricle (LV), and prolonged exposure to an increasing load can lead to LV structural remodeling (8). Myocardial hypertrophy is a compensatory mechanism in response to the remodeled LV, and HF can develop as a consequence of increased LV stiffness and the presence of diastolic dysfunction (9).

Intervention to prevent or attenuate BP rise could lead to substantial reductions in HF development and further morbidity and mortality risks (10). Despite the effectiveness of lifestyle modifications and widespread use of antihypertensive medications, the proportions of HTN awareness, treatment and lifestyle modification are low (11). Therefore, HTN continues to be an important public health challenge worldwide.

Chinese herbal medicine (CHM) has been used for thousands of years to treat or prevent diseases, including HTN. Two systemic reviews recently reported that CHM appeared to have significant antihypertensive effects and may possibly improve outcomes in CVD and HF (12, 13). At present, most population-based studies have focused on the potential antihypertensive effects and prescription patterns of CHM in patients with HTN (14–16). To the best of our knowledge, no studies have investigated the use of CHM for reducing the risk of HF in hypertensive patients. The purpose of this study was to investigate whether the use of CHM can reduce the risk of HF in hypertensive patients. The hypothesis of the study was that the risk of HF would be further reduced by CHM in patients with HTN.

This report presents the first large-scale population-based retrospective cohort study providing evidence of the association between CHM use and HF risk in patients with HTN, which was conducted by analyzing claims data from the NHIRD in Taiwan. The main findings were as follows: (1) CHM use may prevent HF in hypertensive patients because it was associated with a 0.85-fold risk. This effect is related to the number of medication days, since the risk of HF in hypertensive patients who used CHM for more than 180 days/year fell to 0.65 times that in the control group. (2) Compared with controls, male hypertensive patients or those with older age, CHD, CVD or DM had a higher risk of developing HF, but CHM use could not reduce the risk of HF in male hypertensive patients or those with comorbidity. (3) The most common CHM formula prescribed was TMGTY, and the most common single herb prescribed was Dan Shen.

Of the many risk factors for HF, some can be controlled, but not others. Age and sex are two uncontrollable risk factors for HF. In general, the incidence of HF is higher in older than in younger people (18). Simply getting older could increase the risk of HF because of cardiac aging processes including fibrosis, inflammation, mechanical stiffening and diastolic dysfunction, mitochondrial dysfunction, and a growing imbalance between loss and birth of cardiomyocytes (18). Different diagnostic criteria, categorizations, and sources of patients in epidemiological research may produce different results (4). In the Framingham Heart Study, HF was greater in hypertensive women than in men (6). However, recent epidemiological research on HF reported that women have a significantly lower incidence rate of HF compared to men in all age categories except the category of age > 74 years, and HF with preserved ejection fraction is more common in women (4). The data in our study showed that hypertensive patients aged more than 65 years had a 5.3-fold higher risk of developing HF as compared with those aged < 45 years. Consistent with a previous study by Rismiati and Lee (19), our results showed that the risk of developing HF was 1.34-folds higher in male patients with HTN than in female patients with HTN. Our study results revealed that patients treated with CHM after diagnosis of HTN showed a significantly lower risk of HF in the future. In addition, better effects occurred with longer durations of CHM treatment. CHM use can reduce the risk of HF, particularly in hypertensive females and those aged more than 45 years, but not in hypertensive males. These are encouraging results, for they suggest that CHM may be beneficial for cardiac aging or reduce the increased risk of HF caused by aging, which is categorized as an uncontrollable risk factor. The HF risk was not significantly reduced by CHM in patients aged younger than 45 years and in males. The possible explanations include that hypertensive males and those diagnosed with HTN before 45 years old may have unhealthy diets or lifestyles, such as smoking or physical inactivity, obesity, or congenital heart disease (4).

The development of HF has mixed etiologies. Several etiologies, such as CHD, hypertensive heart disease, hyperlipidemia, and DM, may increase the risk of HF (20). Our results showed that patients with HTN who presented with comorbidities, including CHD, CVD, or DM, had higher risk of developing HF, but this trend was not found in those with hyperlipidemia or peripheral arterial occlusive disease. The results showed that CHM use could not reduce the risk of developing HF in hypertensive patients with concomitant diseases mentioned above. A possible reason is that the CVD continuum is a sequence of cardiovascular events, one which begins from several cardiovascular risk factors including DM, hyperlipidemia, HTN, smoking and obesity, followed by CHD or myocardial infarction, and eventually end-stage HF and death (21). Higher numbers of these risk factors are associated with greater chances of developing HF. Previous studies have shown that the prevalence of HTN is higher in low and middle-income countries than in high-income countries (3), and low socioeconomic status independently increases risk of HF (4). Our results showed that the levels of income and urbanization were not associated with higher risk of HF. The possible reasons are the increased awareness, treatment, control and compliance with medication for HTN among Taiwanese people, as well as the improvement of accessibility to medical resources in rural areas. In addition, the results of the study indicated that hypertensive patients with higher incomes or those living in rural areas had lower risk of developing HF while receiving adjuvant CHM treatment. The cause of these findings is not fully understood. One possible explanation is high-income patients who receive adjuvant CHM treatment may tend to seek and can afford more complementary and alternative medicine therapies, as well as having better health behaviors (22). Healthcare resources are often difficult to access in rural areas. Among rural residents, the utilization of CHM as adjuvant therapy with antihypertensive drugs may suggest that they tend to seek and have the ability to pursue better medical care.

According to HTN practice guidelines, the pharmacological treatment options consist of ACEI, ARB, β-blockers, CCB and diuretics (23). A previous study showed that ACEI, ARB, and thiazide diuretics appear to be most effective in preventing new-onset HF in hypertensive patients (24). Our results were not consistent with that previous study, for our study revealed that hypertensive patients who received ARB had lower risk of developing HF; those who received α-blockers, β-blockers, or CCB were not at lower risk for HF; and those who received ACEI or diuretics were at higher risk for HF. The cause of this disparity is unknown, and possible reasons include different study designs, different lifestyles, or genetic factors in other studies. In addition, patients who require treatment with β-blockers or diuretics may have other specific conditions that increase HF risk, such as DM, post myocardial infarction, atrial fibrillation, or fluid overload (23). As for ACEI, which increased HF risk in our study, one possible reason is that side effects such as dry cough may have led to patient non-adherence. Our additional results for Table 1 showed that 41.02% of CHM non-users and 40.78% of CHM users had been prescribed both ACEI and ARB during the follow-up period. We were unable to identify whether they used ACEI and ARB concomitantly or received ARB after an initial treatment with ACEI by NHIRD. However, we believe that the latter is more likely because the combination ACEI and ARB is contraindicated by Taiwan's hypertension guidelines (25). The high proportion of alternative use of ARB following initial treatment of ACEI suggested the possibility of patient non-adherence. It is especially notable that the results demonstrated that CHM use in combination with β-blockers was effective in reducing the risk of developing HF in hypertensive patients as compared with the risk in patients with no CHM use.

The top 10 most frequently prescribed formulae for treating HTN are indicated for a variety of syndromes based on TCM theory, such as TMGTY for the syndrome of internal stirring of liver wind, Xue-Fu-Zhu-Yu-Tang for the syndrome of heart blood stagnation, Gou-Teng-San for the syndrome of liver wind, JWXYS for the syndrome of liver qi stagnation, Zhi-Gan-Cao-Tang for the syndrome of heart qi deficiency, Bu-Yang-Huan-Wu-Tang for the syndrome of qi deficiency and blood stagnation, and Liu-Wei-Di-Huang-Wan and its derivatives for the syndrome of kidney deficiency (14). CHMs have the potential to reduce BP by single or multiple pharmacological mechanisms, such as increasing the nitric oxide (NO) level, blocking the calcium channel, inhibiting the rennin-angiotensin system (RAS), dieresis, and the central mechanism, and they also have anti-cardiac hypertrophic effects by regulating the p38/NF-κb pathway (14, 15). Our results are consistent with those of previous studies (14, 16) in that the most commonly used CHM formula was TMGTY, and the most commonly used single herb was Dan Shen. The combination of TMGTY and Dan Shen was the most common co-prescription for treating HTN (16). TMGTY is usually used to treat HTN-related symptoms in clinical practice. Multiple activities of TMGTY in treating HTN have been demonstrated, including increasing NO levels, blocking calcium channels, regulating the RAS, increasing the diuretic effect, and acting on the central vasomotor center (14, 16). TMGTY has also been reported to lower total cholesterol, prevent the stroke risk in hypertensive patients, and improve clinical symptoms and quality of life (16). Dan Shen has also been reported to have multiple activities in the cardiovascular system, including anti-HTN, anti-atherosclerosis, anti-thrombosis and cardioprotection (16). Moreover, several beneficial effects of Dan Shen in HF prevention include an anti-cardiac hypertrophic effect, reducing adriamycin-induced cardiomyopathy, and improving cardiac angiogenesis and ejection function by modulating the HIF1α/VEGFA signaling pathway after myocardial infarction (15). However, further analysis showed TMGTY and Dan Shen were unable to reduce the risk of HF. The frequently prescribed formula JWXYS and the single herbs Ge Gen, Huang Qi, Du Zhong, Huang Qin, and Chuan Xiong were significantly associated with lower risk of HF. JWXYS is rarely prescribed alone by TCM physicians for treating HTN, however, our study revealed that it was the fourth most commonly prescribed formula. JWXYS is used widely to reduce anxiety, sleep disturbance, and fatigue as well as to manage mood swings associated with climacteric syndrome (26). TCM physicians usually combined it as adjuvant formula to stabilize emotions and restore balance of the autonomic nervous system (27). Patients with such disorders might be at a higher risk of developing HTN, and this may explain the frequent prescription of JWXYS for HTN. Its modified formulation, Xiao-Yao-San (XYS), is beneficial for hypertensive patients in lowering BP, improving depression, regulating blood lipid levels, and inhibiting inflammation (28). So far, no studies have mentioned that JWXYS can prevent HF. This is a new finding that JWXYS can prevent new-onset HF in hypertensive patients. CHM contains numerous active components or phytochemicals that may have multifaceted mechanisms of cardioprotective effects, including antioxidant, vasorelaxant, anti-inflammatory, anti-proliferative, or diuretic effects (29). Five frequently used single herbs, namely, Ge Gen, Huang Qi, Du Zhong, Huang Qin, and Chuan Xiong, were found to reduce the risk of HF in hypertensive patients. The possible mechanisms involved in preventing HF at least include inhibition of angiotensin converting enzyme or angiotensin II receptor (29–32).

The strengths of this study include the comprehensiveness of the nationwide database, which eliminated the possibility of recall bias in questionnaire assessments and ensured that CHM was used before HF occurred. In addition, the study used a multivariate adjusted hazard ratio with a period of 16 years follow-up to confidently estimate the effects of CHM therapy on associations between HTN and the risk of HF over a long latency period. Despite these strengths, several limitations should be noted when interpreting the results of the present study. First, it was difficult to obtain the results of BP levels, body weight measurements, body mass index, medical compliance, lifestyle, and other clinical characteristics, including smoking, alcohol consumption and education level, for individual patients in detail from the NHIRD. The failure to adjust for putative risk factors may have resulted in biased estimates of risks of HF in our sample. Second, other preparations of Chinese herbal remedies, health foods containing natural herbs, folk medicine, and direct purchases from TCM herbal pharmacies are not reimbursed by NHI and thus were not analyzed in this study. However, the high healthcare insurance coverage and low prices of government-approved CHM have led to a reduction in herbal folk medicine use. This missing data could possibly have resulted in underestimation of the cumulative frequencies and may have weakened the effect of the specified CHM, but the effect, if any, should be negligible.

This 16-year follow-up cohort study found that the use of CHM during the treatment of HTN was associated with a 15% lower risk of developing HF compared with the risk among CHM non-users, and the effect increased to a 35% lower risk with a duration of CHM treatment > 180 days. Our study results suggest that the use of CHM as an adjuvant therapy in hypertensive patients may prevent subsequent HF. Further long-term prospective studies or randomized, placebo-controlled clinical trials are needed to validate these findings.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by China Medical University Hospital. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

M-YT and C-TL designed the study and wrote the manuscript. CH, and I-LH acquired the data. K-CH analyzed the data. M-YT and Y-HC revised the manuscript. All authors read and approved the final manuscript.

This study was supported in part by the Taiwan Ministry of Health and Welfare Clinical Trial Center (MOHW 110-TDU-B-212-124004), and China Medical University Hospital (DMR-111-105). We are grateful to the Health Data Science Center, China Medical University Hospital, for providing administrative, technical, and funding support. The funders had no role in the study design, data collection and analysis, the decision to publish, or preparation of the manuscript. No additional external funding was received for this study.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.